CN112783210B - Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system - Google Patents

Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system Download PDF

Info

Publication number
CN112783210B
CN112783210B CN202110005312.1A CN202110005312A CN112783210B CN 112783210 B CN112783210 B CN 112783210B CN 202110005312 A CN202110005312 A CN 202110005312A CN 112783210 B CN112783210 B CN 112783210B
Authority
CN
China
Prior art keywords
unmanned aerial
population
aerial vehicle
chromosomes
chromosome
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110005312.1A
Other languages
Chinese (zh)
Other versions
CN112783210A (en
Inventor
陈盈果
王原
何磊
沈大勇
姚锋
王涛
张忠山
吕济民
陈宇宁
孙文广
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Aerospace Information Research Institute of CAS
Original Assignee
National University of Defense Technology
Aerospace Information Research Institute of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology, Aerospace Information Research Institute of CAS filed Critical National University of Defense Technology
Priority to CN202110005312.1A priority Critical patent/CN112783210B/en
Publication of CN112783210A publication Critical patent/CN112783210A/en
Application granted granted Critical
Publication of CN112783210B publication Critical patent/CN112783210B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
    • G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Abstract

The invention discloses a multi-target control parameter optimization method of an unmanned aerial vehicle cluster control system, which comprises the following steps of 1, acquiring position information and target position information of an unmanned aerial vehicle cluster, and constructing an unmanned aerial vehicle cluster control model based on an artificial potential field; step 2, expanding the unmanned aerial vehicle cluster control model into a rule base control model with a plurality of rules; and 3, optimally selecting parameters of the rule base control model according to four indexes of mortality, aggregation, isotropy and scene completion speed of the unmanned aerial vehicle cluster. According to the invention, through synchronous optimization of a plurality of targets, parameter combinations which do not show comprehensive performance but have good performance on partial optimization targets are prevented from being eliminated prematurely, and finally, the optimization efficiency of parameters is improved. The artificial chromosomes generated by the external storage pool are used for replacing chromosomes with small contribution degrees to population entropy in the population, and the problem that the optimal parameters cannot be obtained due to early convergence in the solving process is solved.

Description

Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system
Technical Field
The invention belongs to the field of intelligent control of robots, and particularly relates to a multi-target control parameter optimization method for an unmanned aerial vehicle cluster control system.
Background
The group unmanned aerial vehicle control is a main research direction of the current unmanned aerial vehicle control, and how to correspondingly control the behavior of an unmanned aerial vehicle cluster through proper rules is mainly researched so as to ensure that the unmanned aerial vehicle cluster can complete the preset cluster behavior. In the research of numerous unmanned aerial vehicle cluster control rules, a control scheme based on virtual physical rules is also taken as a behavior control scheme which is most widely applied. The first article to propose this method is document 1 "Khatib O.real-time architecture for simulations and mobile robotes. int.J.Robot. Res.,1986,5(1): 90-98. In this paper, the authors introduce a conceptual Artificial force Field (APF) for cluster control of unmanned aerial vehicles. An unmanned aerial vehicle cluster controller based on APF design typically meets the following criteria: any drone individual in the cluster is subjected to at least two forces, one an attractive force from the target orientation and one a repulsive force from other individuals or obstacles. The unmanned aerial vehicle cluster controller designed based on the virtual physical rule generally has the following advantages:
(1) the requirement that the unmanned aerial vehicle cluster can display different behaviors can be met by expanding the rule base;
(2) the individual behavior of the unmanned aerial vehicle can be obtained through simple vector operation;
(3) the control model can be quantitatively analyzed by some theoretical tools.
Although extensive research is currently conducted on the unmanned aerial vehicle cluster control model aiming at virtual physical rules, the model still needs to solve a crucial problem so far: how to guarantee that unmanned aerial vehicle control model all can have stable good performance under the environment of difference.
To address this problem, one important class of solutions is to allow the control model to use variable parameters. The research on unmanned aerial vehicle cluster control models using variable parameters is first traced to document 2 "hettiarachhi sd. distributed evaluation for swarm robotics. phd thesis, University of Wyoming, larami, WY, 2007". Based on the unmanned aerial vehicle cluster optimization method, partial researchers develop various unmanned aerial vehicle cluster control model parameter optimization schemes aiming at different unmanned aerial vehicle cluster control purposes, and the unmanned aerial vehicle cluster control model parameter optimization schemes are used for solving the unmanned aerial vehicle control model parameter optimization problems under different environments. Typically as follows: document 3 "Pugh J, Martini A. parallel learning in heterologous multi-robot handover. in Proceedings of the IEEE convergence on evolution calculation 2009, pp: 3839-3846. Piscataway: IEEE Press" uses an optimization method based on a particle swarm algorithm for solving parameters in the unmanned aerial vehicle cluster control model, and obtains an algorithm performance having better population diversity than a genetic algorithm. Document 4 "Folino G, Forestero A, Spezzzano G.A adaptive flooding for performing adaptive aggregation. Inf.Sci.,2009,179(18): 3059-. The algorithm has better performance aiming at unmanned aerial vehicle clusters of different scales. Document 5 "Yang H, Huang l.generating collecting movement of self-propagating agents by distance-based influence.phys.rev.2014,89(3): 032813." improves the unmanned aerial vehicle cluster control model based on the classical Vicsek model to complete the unmanned aerial vehicle aggregation task. In this study, a variable parameter was used to control the ability of different individuals in a cluster to interact with each other. Document 6 "Zhao M, Su H, Wang M, et al.a weighted adaptive-altitude self-organizing model and its high predicted performance. neuro-output 2016,216(C): 402-. However, these algorithms have the disadvantages that:
in a classical unmanned aerial vehicle control cluster parameter optimization method, when optimization targets are considered, methods such as weighted aggregation are generally adopted, and the problem is finally summarized to be a single-target optimization problem or only one unmanned aerial vehicle control problem is solved. However, in the process of executing tasks by the unmanned aerial vehicle cluster, a conflict relationship exists between the aggregative requirement of the unmanned aerial vehicle cluster and the requirement for avoiding mutual collision of the unmanned aerial vehicles, and optimization of the cluster as a single target problem does not really solve multiple target problems in the unmanned aerial vehicle cluster, such as the speed requirement for completing a target, the aggregative requirement of the unmanned aerial vehicle cluster, the requirement for avoiding mutual collision of the unmanned aerial vehicle cluster, and the like, which have potential conflict target problems, so that the unmanned aerial vehicle cluster cannot be well controlled. Therefore, in order to better control the unmanned aerial vehicle cluster, a parameter optimization control method for the unmanned aerial vehicle cluster is urgently needed, so that various target problems of the unmanned aerial vehicle cluster can be controlled.
Disclosure of Invention
The invention aims to solve the technical problem of rapidly and better controlling an unmanned aerial vehicle cluster and provides a multi-target control parameter optimizing method of an unmanned aerial vehicle cluster intelligent control system.
In order to solve the problem, the technical scheme adopted by the invention is as follows:
a multi-target control parameter tuning method for an unmanned aerial vehicle cluster control system comprises the following steps:
step 1: acquiring scene information of an unmanned aerial vehicle cluster, and constructing an unmanned aerial vehicle cluster control model based on an artificial potential field;
the speed of the ith unmanned aerial vehicle individual at the next moment t +1 is
vi(t+1)＝vi(t)+ΔviWherein
Figure GDA0003474671990000031
Wherein
Figure GDA0003474671990000032
Figure GDA0003474671990000033
Is the sum of all repulsive forces from other drones j received by the current drone i, and N is the number of all drones in the drone cluster,
Figure GDA0003474671990000034
maximum value of the repulsive force range;
Figure GDA0003474671990000035
Figure GDA0003474671990000036
the representative drone i receives the repulsive force, r, of another drone jijIs the distance between drone i and drone j. p is a radical ofiAnd pjThe positions of unmanned plane i and unmanned plane j in the current space are represented by vectors;
Figure GDA0003474671990000037
Figure GDA0003474671990000038
for the magnitude of equidirectional force, v, that unmanned aerial vehicle i receivesjIs the current unmanned aerial vehicle's speed, n, expressed by a vectoraliIs located at R1The number of drones within the range,
Figure GDA0003474671990000039
is R1Maximum value of (1), R1The radius of the equidirectional force range.
Figure GDA00034746719900000310
Wherein the content of the first and second substances,
Figure GDA00034746719900000311
is the gravitation that the unmanned aerial vehicle i receives at a certain moment.
Figure GDA00034746719900000312
The maximum value of the gravity range.
Figure GDA00034746719900000313
Figure GDA00034746719900000314
Representing the attraction released by a single drone j to drone i,
Figure GDA00034746719900000315
is R2Is measured. R2Radius, R, of the maximum detection range of the unmanned aerial vehicle, which is also the gravity range0＝1/3R2,R1＝2/3R2，R0Is a radius of a range of the repulsive force,
Figure GDA00034746719900000316
wherein the content of the first and second substances,
Figure GDA0003474671990000041
indicating the magnitude of the directional force, p, currently experienced by drone itRepresenting the target area position, p, of all drones represented by a vectoriFor the position where drone i is currently located, r, represented by a vectoritIs the distance between the destination and the ith unmanned aerial vehicle.
Figure GDA0003474671990000042
Figure GDA0003474671990000043
For the collision avoidance force to which the unmanned aerial vehicle i is subjected,
Figure GDA0003474671990000044
distance r of the unmanned plane furthest capable of detecting the obstacleimMinimum distance, p, from the surface of the obstacle detected by the drone to the current individualmM is the total number of obstacles that can be detected by the current individual, which is the spatial position of the obstacle M represented by the vector.
Step 2: expanding the unmanned aerial vehicle cluster control model into a rule base control model with a plurality of rules;
Figure GDA0003474671990000045
wherein, rule 1 to 4 are respectively:
rule 1 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion force range and no other obstacle exists in the detection range;
rule 2 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and no other obstacles exist in the detection range;
rule 3 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion range and other obstacles exist in the detection range;
rule 4 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and other obstacles exist in the detection range;
parameter matrix
Figure GDA0003474671990000046
Parameters of the rule base control model;
and step 3: and taking the mortality, the aggregative property, the isotropy and the scene completion speed of the unmanned aerial vehicle cluster into consideration, and optimally selecting the parameters of the rule base control model to obtain the optimal unmanned aerial vehicle cluster control model to realize the control of the unmanned aerial vehicle cluster.
The invention also provides a multi-target control parameter tuning system of the unmanned aerial vehicle cluster control system, which comprises a memory and a processor, wherein the memory stores a computer program of the multi-target control parameter tuning method of the unmanned aerial vehicle cluster control system, and the processor realizes the steps of the method when executing the computer program.
Compared with the prior art, the invention has the following beneficial effects:
(1) on the basis of the traditional parameter optimization scheme of the unmanned aerial vehicle cluster control system, the invention adopts an improved pareto selection multi-objective optimization method aiming at the requirements of multi-angle and multi-direction optimization required by the unmanned aerial vehicle cluster control model, and realizes multi-objective synchronous optimization of parameter optimization of the unmanned aerial vehicle cluster control system, thereby solving the problem that a corresponding customized optimization scheme can be flexibly given according to different scene requirements and user preferences in the engineering problem. Aiming at the situation that conflicts may exist among different optimization targets in many engineering problems, a plurality of targets are synchronously optimized, and the condition that some parameter combinations which do not show the comprehensive performance but have good performance on part of the optimization targets are eliminated too early can be avoided, so that the optimization efficiency of the parameters is finally improved.
(2) Aiming at the problem of algorithm precocity existing in the problem of parameter optimization of an unmanned aerial vehicle cluster control system, population diversity is detected, and for populations with undersized population diversity, artificial chromosomes generated by chromosomes stored in an external storage pool are used for replacing chromosomes with small contribution degrees to population entropy in the populations, so that the diversity of the populations is improved, and further the problem that in the problem of actual engineering, due to the fact that the population diversity is insufficient, the algorithm is too early converged, and the optimal optimization effect on a control parameter set cannot be achieved is solved.
(3) Compared with a classical unmanned aerial vehicle cluster control system, the unmanned aerial vehicle cluster control system optimized by the algorithm described by the invention is proved to have great improvement on three measurement indexes of individual aggregation, individual collision avoidance performance and target speed completion under the scene that a cluster passes through obstacles.
Drawings
FIG. 1 is a system flow diagram of the present invention;
FIG. 2 illustrates the detection range of various forces;
FIG. 3 illustrates a population coding scheme;
FIG. 4 is a flow chart of an improved genetic algorithm;
FIG. 5 is a simulation scenario presentation;
FIG. 6 is a simulation result of the control system of the present invention;
fig. 7 shows simulation results of the Reynolds control model.
Detailed Description
Fig. 1 to 7 show a multi-target control parameter tuning method of an unmanned aerial vehicle cluster control system, which includes the following steps, as shown in fig. 1:
step 1: acquiring scene information and target position information of an unmanned aerial vehicle cluster, and constructing an unmanned aerial vehicle cluster control model based on an artificial potential field; the scene information of the unmanned aerial vehicle cluster in this embodiment includes the scale of the unmanned aerial vehicle cluster, the maximum speed of the unmanned aerial vehicle, the maximum detection range, the maximum communication range and other information.
The speed of the ith unmanned aerial vehicle individual at the next moment t +1 is
vi(t+1)＝vi(t)+ΔviWherein
Figure GDA0003474671990000061
Wherein { a, b, c, d, e } is five different force weights, which are the objects of the present invention to be optimized.
Figure GDA0003474671990000062
Figure GDA0003474671990000063
Is the sum of all repulsive forces from other individuals suffered by the current unmanned aerial vehicle individual i, N is the set of all unmanned aerial vehicles in the unmanned aerial vehicle cluster,
Figure GDA0003474671990000064
maximum value of the repulsive force range;
Figure GDA0003474671990000065
Figure GDA0003474671990000066
representing the unmanned individual i being subjected to the repulsive force, r, of another individual jijIs the distance, p, between individual i and individual j of the droneiAnd pjThe positions of the individual i and the individual j in the current space are represented by vectors;
Figure GDA0003474671990000067
Figure GDA0003474671990000068
magnitude, v, of equidirectional force applied to individual i of unmanned aerial vehiclejIs the current unmanned aerial vehicle's speed, n, expressed by a vectoraliIs located at R1Number of drones within range, R as shown in FIG. 21Is the radius of the same-direction force range,
Figure GDA0003474671990000069
is R1Maximum value of (d);
Figure GDA00034746719900000610
wherein the content of the first and second substances,
Figure GDA00034746719900000611
the gravity of the unmanned aerial vehicle individual i at a certain moment,
Figure GDA00034746719900000612
Figure GDA00034746719900000613
representing the attractive force released by the drone individual j on the individual i,
Figure GDA00034746719900000614
is R2Maximum value of (1), R2Radius, R, of the maximum detection range of the unmanned aerial vehicle, which is also the gravity range0＝1/3R2,R1＝2/3R2，R0Radius of the repulsive force range, R1The radius of the equidirectional force range.
Figure GDA0003474671990000071
Wherein the content of the first and second substances,
Figure GDA0003474671990000072
indicating the magnitude of the directional force, p, currently experienced by drone itRepresenting the target area position, p, of all drones represented by a vectoriFor the position where drone i is currently located, r, represented by a vectoritIs the distance between the destination and the ith unmanned aerial vehicle.
Figure GDA0003474671990000073
Figure GDA0003474671990000074
For the collision avoidance force to be applied to the unmanned aerial vehicle individual i,
Figure GDA0003474671990000075
the distance at which the individual is the farthest detectable to an obstacle, rimMinimum distance, p, from the surface of the obstacle detected by the drone to the current individualmThe space position where the obstacle M is represented by a vector is, and M is the total number of obstacles which can be detected by the current individual;
step 2: expanding the unmanned aerial vehicle cluster control model into a rule base control model with a plurality of rules;
Figure GDA0003474671990000076
wherein, rule 1 to 4 are respectively:
rule 1 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion force range and no other obstacle exists in the detection range;
rule 2 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and no other obstacles exist in the detection range;
rule 3 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion range and other obstacles exist in the detection range;
rule 4 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and other obstacles exist in the detection range;
parameter matrix
Figure GDA0003474671990000077
Parameters of the rule base control model;
and step 3: and taking the mortality, the aggregative property, the isotropy and the scene completion speed of the unmanned aerial vehicle cluster into consideration, and optimally selecting the parameters of the rule base control model to obtain the optimal unmanned aerial vehicle cluster control model to realize the control of the unmanned aerial vehicle cluster.
The mortality index refers to the capability of keeping no collision between unmanned aerial vehicles in a formation obstacle crossing scene, and is expressed as
Figure GDA0003474671990000081
Wherein r isdeathFor mortality, NdeathIs the number of the unmanned aerial vehicle individuals who die after the task is completed. N is the total number of drones. In this embodiment, the death of the drones means that two drones collide (the distance is 0).
The aggregative indicator refers to the ability of the drone to maintain the flight of the formation, and is expressed as:
Figure GDA0003474671990000082
wherein the content of the first and second substances,
Figure GDA0003474671990000083
to be the centroid of the drone cluster at time t,
Figure GDA0003474671990000084
for the actual position of UAV i at time t, NT is the total simulation duration, daggThe aggregation index of the unmanned aerial vehicle cluster.
The isotropy index is an index for keeping the flight isotropy in the unmanned aerial vehicle formation, and the isotropy index is expressed as:
Figure GDA0003474671990000085
where
Figure GDA0003474671990000086
in this formula, vscIs an index of the isotropy,
Figure GDA0003474671990000087
for the direction of flight of drone i at time t, δ in degreestIs the flight direction of the cluster of drones at time t, expressed in degrees.
The scene completion speed refers to the average time required for each unmanned aerial vehicle in the unmanned aerial vehicle cluster to start from a scene starting point and finally reach a scene end point in the unmanned aerial vehicle cluster task scene. The scene completion speed index is expressed as:
Figure GDA0003474671990000088
in the formula
Figure GDA0003474671990000089
Is a scene completion speed indicator, TCiThe time required for drone i to reach the scene end.
The method for optimizing and selecting the parameters in the embodiment is an Improved reinforced Pareto evolution Algorithm SPEA2(Improved Strength Pareto evolution Algorithm 2, ISPEA2) multi-objective optimization Algorithm.
An improved enhanced pareto evolution algorithm ISPEA2 multi-objective optimization algorithm is shown in fig. 4, and specifically includes:
step 3.1: a population code, which initializes the population, as shown in fig. 3, wherein the population code adopts a one-dimensional sequence code; one locus in each chromosome represents one of the parameters to be optimized;
step 3.2: the population diversity D (P) was measured. If the population diversity is less than a fixed threshold dlowStep 3.7 is executed, otherwise step 3.3 is executed;
in this embodiment, the measurement of the population diversity d (p) can be performed by the following formula:
Figure GDA0003474671990000091
in this formula, | L | is the length of the solution space diagonal, and in the present invention, | L | is the length of the solution space diagonal
Figure GDA0003474671990000092
ps is the number of chromosomes in the population, sijIs the value at the jth gene position on the ith chromosome in the population,
Figure GDA0003474671990000093
is the mean value at the jth locus on all chromosomes in the population. By this formula, the difference d (p) of chromosomes in the entire population can be measured.
Step 3.3: selecting chromosomes in the population by adopting a non-dominance rule-based method according to a mechanism of strengthening the pareto SPEA2 algorithm;
in this embodiment, different from the conventional scheme of converting a multi-target problem into a single target for solving in a weighted aggregation manner, the solution set obtained finally can be guaranteed to have better diversity by solving the multi-target optimization problem with a non-dominated rule. To introduce the rules of non-dominance, the concept of Pareto Domination (Pareto Domination) is first introduced. The pareto domination means that in the process of solving the multi-objective problem, one solution a is superior to a solution B in the performance of all evaluation indexes, and is called a solution a dominating (Dominate) solution B. In the Solution process of the problem faced by the present invention, in the chromosome selection step of each iteration, the population of each generation is firstly sorted based on the Non-Dominated rule, and if one Solution is not Dominated by any other Solution, the Solution is called as the Non-Dominated Solution (Non-Dominated Solution) of the current step. If a solution is dominated by at least one solution in the current solution set, the solutions in the population are sorted according to the number of other solutions that dominate the solution (assuming that if a solution a is dominated by only one solution B in the current solution set, the dominated number of solutions a is 1). By the above method, the population obtained for each generation can be sorted, and the chromosome selection operation is performed as follows: in the ISPEA2 algorithm, before each iteration starts, an external population Pool (matting Pool) is firstly set for generating a population entering the next generation, and the external population Pool of the t generation is set as
Figure GDA0003474671990000094
Let the tth generation
Figure GDA0003474671990000095
The resulting population is pt. To generate
Figure GDA0003474671990000096
Firstly, the following components are mixed
Figure GDA0003474671990000097
And ptAll non-dominant solution additions in
Figure GDA0003474671990000098
Then, to
Figure GDA0003474671990000099
The size of (2) is judged. If it is not
Figure GDA00034746719900000910
The number of chromosomes in the chromosome is greater than a predetermined number
Figure GDA00034746719900000911
Then the reduction is done by a population density based method (see literature 7 "Ziegler, E., Laumans, M., and Thiele, L. (2001): SPEA2: Improving the string of Pareto evolution algorithm for multiple optimization. in Proceedings of evolution methods for design, and control, Barcelona, Spain (pp.95-100.))
Figure GDA00034746719900000912
Up to the number of chromosomes in
Figure GDA00034746719900000913
Number of chromosomes in is equal to
Figure GDA00034746719900000914
If the number of chromosomes in the population is less than
Figure GDA00034746719900000915
Then the same population density based selection method will be used
Figure GDA00034746719900000916
And ptThe solution in the dominant solution is added according to the order of the dominant number from small to large
Figure GDA0003474671990000101
Up to
Figure GDA0003474671990000102
Number of chromosomes in is equal to
Figure GDA0003474671990000103
The chromosome selection is carried out by using the non-dominant rule, more chromosomes which may not be retained under the single target rule can be retained, for example, when the chromosome selection is carried out by using the non-dominant rule, individuals with better aggregative indexes but slower target completing speed and individuals with faster completing speed but poorer aggregative property can be retained at the same time; but when converted to a single target, it is possible that one of them will be eliminated to achieve simultaneous optimization for different optimization targets.
Step 3.4: performing cross operation on the population; randomly swapping a part of the gene locus between two chromosomes;
step 3.5: carrying out mutation operation on the population; randomly selecting at most three gene positions in a chromosome, and randomly generating other numerical values to replace numerical values on the gene positions in the selected chromosome so as to realize the variation of the chromosome;
step 3.6: judging whether a stopping condition is reached, if not, collecting the known pareto solution of the current generation and returning to the step 3.2, otherwise, finishing the algorithm and outputting a result; in this embodiment, the termination condition is that the iteration number reaches a preset upper limit.
Step 3.7: generating artificial chromosomes from chromosomes stored in an external storage pool;
in this embodiment, the method of generating the artificial chromosome is:
step 3.7.1: calculating the contribution degree of each chromosome in the current population to the current population entropy,
in the embodiment, by means of the concept of entropy in informatics, the contribution degree of each chromosome to population entropy is calculated, and the contribution of the chromosome to population diversity is measured. It should be noted that, since the entropy cannot be directly calculated in the normal case, we indirectly calculate the contribution degree of a specific chromosome to the population by calculating the difference between the current entropy of the population and the entropy after removing the specific chromosome in the population. By calculating the contribution degree of each chromosome in the population to the current population entropy, chromosomes with less contribution degree to the population entropy in the population can be eliminated, and new artificial chromosomes are injected to improve the population diversity.
First, the probability of the value v appearing on the chromosome locus j in the current population is calculated:
Figure GDA0003474671990000104
wherein, PvjIs the probability of the occurrence of the value v on the factor j, XvjThe number of times that the value v is distributed to the gene position j in all the chromosomes of the current population after statistics;
secondly, calculating the contribution degree of the gene position j on the chromosome to the population entropy;
Figure GDA0003474671990000105
v is the complete set of values V;
step 3.7.2: calculating the entropy of the whole population;
Figure GDA0003474671990000111
wherein the PCDpAs entropy of the entire population, NpFor the dimension of the problem, N in this embodimentpRefers to the number of parameters;
step 3.7.3: calculating the contribution degree of each chromosome i in the current population to the entropy of the whole population,
CD(i)＝PCDp-PCDp-i
in the above formula, CD (i) is the contribution degree of chromosome i to the entropy of the whole population, PCDpFor entropy of the entire population, PCDp-iRemoving the staining for the entire populationEntropy of population after body i. Through the steps, the contribution degree of each chromosome in the population to the entropy of the current population can be calculated.
Step 3.7.4: chromosomes used to generate artificial chromosomes are collected in an external storage pool: collecting the non-dominant solution set generated in the step at each step of iteration and adding the solution set into an external storage pool, when the number of chromosomes in the external storage pool is less than a preset threshold value
Figure GDA0003474671990000112
When all non-dominant solutions are placed in the external pool, when the number of chromosomes in the external pool is greater than that
Figure GDA0003474671990000113
And sequencing the chromosomes in the external storage pool according to the contribution degree of the chromosomes to the population entropy, and removing the individuals with low contribution degree to the population entropy out of the storage pool until the number of the chromosomes in the external storage pool is equal to that in the external storage pool
Figure GDA0003474671990000114
In the present invention, setting
Figure GDA0003474671990000115
Step 3.7.5: and respectively counting the probability of each numerical value v appearing on each gene position of all chromosomes in the external storage pool, and assigning values to each gene position on the artificial chromosome according to the probability value when the artificial chromosome is generated until all required artificial chromosomes are generated. In this example, the amount of the artificial chromosome produced by the preliminary measurement was
Figure GDA0003474671990000116
It is more suitable.
According to this method, the higher the probability that a value v appears at the corresponding locus j in the external pool, the greater the probability that the value v is assigned to the corresponding locus on the newly created artificial chromosome.
Step 3.8: and replacing the chromosome with smaller contribution degree to population entropy in the current population by the artificial chromosome and executing the step 3.4.
And 4, step 4: and (4) selecting a solution. After optimization by the ISPEA2 algorithm, a set of pareto solution sets will be generated. The solution set represents a plurality of optimal solutions generated under the condition of simultaneously considering a plurality of different optimization targets faced by the invention. Considering that in engineering problems, one solution in a group of solutions is required to be selected as a control parameter which is actually used in a targeted manner, the invention designs a linear weighted aggregation strategy based on the preference of a decision maker for generating the final optimal solution. The specific formula is as follows:
Figure GDA0003474671990000121
the formula represents: first, considering that different optimization objectives may have different dimensions, the different optimization objectives are first normalized. Then, for different decision maker preferences, corresponding weights { w ] are designed for four different optimization objectives1,w2,w3,w4}. Finally, the optimal solution is obtained by means of weighted aggregation. In the formula, in the above-mentioned formula,
Figure GDA0003474671990000122
representing the mortality rate of the current solution,
Figure GDA0003474671990000123
being the minimum of all solutions output by the algorithm,
Figure GDA0003474671990000124
the maximum value of the mortality rate in all solutions output by the algorithm;
Figure GDA0003474671990000125
represents an aggregative indicator of the current solution,
Figure GDA0003474671990000126
is the minimum value of the aggregative indicator in all solutions output by the algorithm,
Figure GDA0003474671990000127
the maximum value of the aggregative indicator in all solutions output by the algorithm;
Figure GDA0003474671990000128
representing the co-directionality factor of the current solution,
Figure GDA0003474671990000129
is the minimum value of the isotropy index in all solutions output by the algorithm,
Figure GDA00034746719900001210
the maximum value of the isotropy index in all solutions output by the algorithm;
Figure GDA00034746719900001211
representing the scene completion speed of the current solution,
Figure GDA00034746719900001212
is the minimum value of the scene completion speed index in all solutions output by the algorithm,
Figure GDA00034746719900001213
the maximum value of the scene completion speed index in all solutions output by the algorithm. w is a1,w2,w3,w4Respectively, the importance weights of the corresponding targets according with the preference of the decision maker.
The invention also provides a multi-target control parameter tuning system of the unmanned aerial vehicle cluster control system, which comprises a memory and a processor, wherein the memory stores a computer program of the multi-target control parameter tuning method of the unmanned aerial vehicle cluster control system, and the processor realizes the steps of the method when executing the computer program.
The effects achieved by the present invention will be described below by specific experiments.
The specific embodiments of the present invention can be illustrated by one case as follows:
(1) simulation environment: all the algorithms and simulation experiments described in the invention are operated in one use
Figure GDA00034746719900001214
CoreTMi5-4460(3.2GHz), 4GB internal memory.
(2) Basic information of unmanned aerial vehicle formation:
each algorithm in each scene was run 10 times;
the side length of the field is 250 meters;
the maximum speed of the unmanned aerial vehicle is set to be 5 m/s;
the scale of the unmanned aerial vehicle cluster is 20;
the communication delay is 1 second (namely 1 step in the simulation experiment);
considering that the communication range of the unmanned aerial vehicle usually has certain fluctuation in the practical problem, the maximum communication range of a single unmanned aerial vehicle is set to be a random value in the range of 29-31 meters;
also considering the uncertainty of the load on board the drone in the real-world problem, the maximum detection range of a single drone is set to be a random number in the range of 95-105 meters.
(3) And (3) simulation scene display: in this case, four common operating scenarios of the unmanned aerial vehicle clusters are mainly included, and in the four scenarios, the unmanned aerial vehicle clusters all need to reach the arrival area at the upper right corner from the departure position at the lower left corner of the figure under the precondition that formation flight is kept (that is, the distance between the unmanned aerial vehicle clusters does not exceed a certain upper limit). The simulation scenario is shown in fig. 5. Fig. 5 is, from top left to bottom right: a non-obstacle scene, a convex obstacle scene, a channel obstacle scene, and a concave obstacle scene. These four scenarios are often used to evaluate the effectiveness of the drone cluster control system.
(4) The comparison method comprises the following steps: in the invention, two different comparative experiments are adopted to determine the effectiveness of the algorithm. First, for comparing the effectiveness of the algorithm, the optimization method described in the present invention and two other widely used multi-objective optimization algorithms (NSGA II, document 7 "Deb, K., Pratap, A., Agarwal, S.and Meyarrivan, T. (2002) 'A fast and electronic multi-objective genetic algorithm: NSGA-II,' in IEEE Transactions on evolution calculation, vol.6, No.2, pp.182-197. 'and document (8)' Zitzler, E., Laumans, M., and Thiele, L. (2001) SPEA2: analyzing the structural parameters evaluation algorithm for optimization, in Proceedings of Analysis, Analysis of the median of the optimization algorithm, and comparison of the effectiveness of the optimization algorithms, and comparison of the optimization algorithms (95) are combined and Analysis of the poor stability of the optimization algorithms, respectively are performed and the comparison of the results of the optimization algorithms and the stability of the optimization algorithms are analyzed and the poor comparison of the optimization algorithms (100), the optimized model is compared with a Reynolds control model widely applied in the field of unmanned aerial vehicle cluster control at present, and the advantages and disadvantages of the control model are analyzed.
(5) The specific implementation scheme is as follows:
in the present invention, the maximum number of generations of operation of the algorithm is set to 100 generations, the population size is set to 30, and the size of the external storage pool is set
Figure GDA0003474671990000131
Again 30.
The method comprises the steps of analyzing the efficiency, optimizing cluster model parameters of the unmanned aerial vehicle by using an improved multi-objective optimization algorithm, comparing the multi-objective optimization algorithm with widely used multi-objective optimization algorithms NSGA II and SPEA2), and evaluating different algorithms in the aspects of mean value, optimal/worst condition, algorithm stability, pareto solution coverage capability and the like. The chromosome coding, population initialization, crossover and mutation operators in the NSGA II and SPEA2 algorithms are the same as those described in the present invention. It should be additionally noted that, because the result of the multi-objective heuristic algorithm has a certain randomness, each scene is run 30 times and all the results are recorded. The final comparison results are shown below:
TABLE 1 pareto frontier solution analysis
Figure GDA0003474671990000141
In the table, DR represents mortality, AI represents aggregative indicator, CI represents homeotropic indicator, and TC represents scene completion speed. It should be noted that the data in table 1 are all subjected to appropriate normalization processing for visually demonstrating the differences between the results of different algorithms. As can be seen from the above table, in a general situation, the algorithm of the present invention has a higher improvement in the worst case of each index compared to other optimization algorithms, and also has a greater advantage in terms of mean and optimum values.
To further analyze the superiority of the algorithm in multi-objective optimization, we introduced a second analysis: Hyper-Volume Analysis. The method is used for analyzing the proportion of a solution set which can be covered by the pareto solution of the multi-objective optimization algorithm in a solution space. A relevant explanation of Hyper-Volume Analysis is found in the document 9 "Ziegler, E., & Thiele, L. (1999). Multi objective evolution algorithms: a comparative case study and the term h space project IEEE Transactions on evolution calculation, 3 (4);" 257- "271". The superiority of the parameter adjustment scheme described in the invention is further illustrated by the Hyper-Volume Analysis result between different algorithms. The invention uses Hyper-Volume Analysis program from "https:// ww2.mathworks. cn/matlabenterral/filexchange/30785-Hyper-Volume-calculation" in performing Hyper-Volume Analysis. The selected reference point is a lower bound {0,0,0,0}, and the upper bound is the maximum value of each optimization index found in each scene by the three algorithms. The Analysis program adopts a Monte Carlo method to carry out Hyper-Volume Analysis, and the invention sets that each operation of each scene is sampled for 50 ten thousand times. The specific results are as follows:
TABLE 2 Hyper-Volume analysis
ISPEA2 SPEA2 NSGAII
HV Difference in HV Difference in HV Difference in
Convex obstacle Mean value 0.42 0.00 0.40 0.05 0.39 0.08
Maximum of 0.44 0.00 0.43 0.02 0.41 0.07
Minimum size 0.40 0.00 0.37 0.08 0.39 0.03
Foveal disorder Mean value 0.38 0.00 0.36 0.05 0.36 0.05
Maximum of 0.41 0.00 0.38 0.07 0.38 0.07
Minimum size 0.35 0.00 0.34 0.03 0.34 0.03
Channel Mean value 0.32 0.00 0.31 0.03 0.30 0.06
Maximum of 0.35 0.00 0.34 0.03 0.33 0.06
Minimum size 0.28 0.00 0.26 0.07 0.27 0.04
Without obstacle Mean value 0.17 0.00 0.17 0.00 0.16 0.06
Maximum of 0.20 0.00 0.19 0.05 0.19 0.05
Minimum size 0.14 0.00 0.14 0.00 0.13 0.07
By combining the results of Hyper-Volume Analysis, the algorithm result described by the invention has better coverage and can cover more solution space, thereby proving that the efficiency performance of the parameter adjusting method described by the invention is better compared with other types of adjusting methods.
Carrying out comparative analysis on the optimized model efficiency
In this section, the optimization model described in the present invention is compared with the Reynolds control model widely used for cluster control of unmanned aerial vehicles at present, and the advantages and disadvantages of the related optimization targets are analyzed, and a trajectory graph analysis is given, as shown in fig. 6 and 7. Compared with the control system widely applied at present, the unmanned aerial vehicle cluster control system optimized by the control system parameter optimization method has better performance in the aspects of unmanned aerial vehicle aggregation, collision probability reduction and task completion speed improvement.
It should be noted that the present invention uses a weighted aggregation method that takes into account the preference of the decision maker when selecting the final control parameters. In this experiment, four weights { w ] in the formula for selecting the optimal solution are set1,w2,w3,w4And the four indexes are all equal to 1, namely, the importance degrees among the indexes are equal. The simulation result is finally obtained as follows:
TABLE 3 controller Performance
Figure GDA0003474671990000151
Figure GDA0003474671990000161
The results show that the control system after parameter optimization of the invention has improved evaluation of each key index compared with the control system widely used at present, which comprises: the mortality rate is reduced by 25%, the aggregative property is improved by 13.9%, the isotropy is improved by 16.8%, and the scene completion time is shortened by 3.5% (average of four scenes). Therefore, it can be shown that the parameter optimization method described in the present invention plays a certain role in improving the efficiency of the unmanned aerial vehicle cluster control system.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (7)

1. A multi-target control parameter tuning method of an unmanned aerial vehicle cluster control system is characterized by comprising the following steps:
step 1: acquiring position information and target position information of an unmanned aerial vehicle cluster, and constructing an unmanned aerial vehicle cluster control model based on an artificial potential field;
the speed of the ith unmanned aerial vehicle individual at the next moment t +1 is
vi(t+1)＝vi(t)+Δvi，
Wherein the content of the first and second substances,
Figure FDA0003474671980000011
Figure FDA0003474671980000012
Figure FDA0003474671980000013
is the sum of all repulsive forces from other individuals suffered by the current unmanned plane individual i, N is the number of all unmanned planes in the unmanned plane cluster,
Figure FDA0003474671980000014
maximum value of the repulsive force range;
Figure FDA0003474671980000015
Figure FDA0003474671980000016
representing the unmanned individual i being subjected to the repulsive force, r, of another individual jijIs the distance, p, between individual i and individual j of the droneiAnd pjThe positions of the individual i and the individual j in the current space are represented by vectors;
Figure FDA0003474671980000017
Figure FDA0003474671980000018
magnitude, v, of equidirectional force applied to individual i of unmanned aerial vehiclejIs the current unmanned aerial vehicle's speed, n, expressed by a vectoraliIs located at R1Number of unmanned aerial vehicles within range, R1Is the radius of the same-direction force range,
Figure FDA0003474671980000019
is R1Maximum value of (d);
Figure FDA00034746719800000110
wherein the content of the first and second substances,
Figure FDA00034746719800000111
the gravity of the unmanned aerial vehicle individual i at a certain moment,
Figure FDA00034746719800000112
Figure FDA0003474671980000021
representing the attractive force released by the drone individual j on the individual i,
Figure FDA0003474671980000022
is R2Maximum value of (1), R2The radius of the maximum detection range of the unmanned aerial vehicle and the radius of the gravity range,
Figure FDA0003474671980000023
R0radius of the repulsive force range, R1Radius of the same direction force range;
Figure FDA0003474671980000024
wherein the content of the first and second substances,
Figure FDA0003474671980000025
indicating the magnitude of the directional force, p, currently experienced by drone itRepresenting the target area position, p, of all drones represented by a vectoriFor the position where drone i is currently located, r, represented by a vectoritThe distance between the destination and the ith unmanned aerial vehicle;
Figure FDA0003474671980000026
Figure FDA0003474671980000027
for the collision avoidance force to be applied to the unmanned aerial vehicle individual i,
Figure FDA0003474671980000028
the distance at which the individual is the farthest detectable to an obstacle, rimMinimum distance, p, from the surface of the obstacle detected by the drone to the current individualmThe space position where the obstacle M is represented by a vector is, and M is the total number of obstacles which can be detected by the current individual;
a. b, c, d and e respectively represent the weight values of the five different forces;
step 2: expanding the unmanned aerial vehicle cluster control model into a rule base control model with a plurality of rules;
Figure FDA0003474671980000029
wherein, rule 1 to 4 are respectively:
rule 1 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion force range and no other obstacle exists in the detection range;
rule 2 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and no other obstacles exist in the detection range;
rule 3 is: when the unmanned aerial vehicle can detect that no other unmanned aerial vehicle exists in the repulsion range and other obstacles exist in the detection range;
rule 4 is: when the unmanned aerial vehicle can detect that other unmanned aerial vehicles exist in the repulsion force range and other obstacles exist in the detection range;
parameter matrix
Figure FDA0003474671980000031
Parameters of the rule base control model;
and step 3: and taking the mortality, the aggregative property, the isotropy and the scene completion speed of the unmanned aerial vehicle cluster into consideration, and optimally selecting the parameters of the rule base control model to obtain the optimal unmanned aerial vehicle cluster control model to realize the control of the unmanned aerial vehicle cluster.
2. The method of claim 1, wherein the method for optimizing the parameters in step 3 is an improved enhanced pareto evolutionary algorithm ISPEA 2.
3. The method of claim 2, wherein the parameter optimization result of the drone cluster control model obtained according to the improved enhanced pareto evolution algorithm ISPEA2 is a set of pareto solution sets; the method further comprises the step 4: and selecting a solution, wherein the method for selecting the optimal solution from the pareto solution set comprises the following steps:
Figure FDA0003474671980000032
Figure FDA0003474671980000033
representing the mortality rate of the current solution,
Figure FDA0003474671980000034
being the minimum of all solutions output by the algorithm,
Figure FDA0003474671980000035
the maximum value of the mortality rate in all solutions output by the algorithm;
Figure FDA0003474671980000036
represents an aggregative indicator of the current solution,
Figure FDA0003474671980000037
is the minimum value of the aggregative indicator in all solutions output by the algorithm,
Figure FDA0003474671980000038
the maximum value of the aggregative indicator in all solutions output by the algorithm;
Figure FDA0003474671980000039
representing the co-directionality factor of the current solution,
Figure FDA00034746719800000310
is the minimum value of the isotropy index in all solutions output by the algorithm,
Figure FDA00034746719800000311
the maximum value of the isotropy index in all solutions output by the algorithm;
Figure FDA00034746719800000312
representing the scene completion speed of the current solution,
Figure FDA00034746719800000313
is the minimum value of the scene completion speed index in all solutions output by the algorithm,
Figure FDA00034746719800000314
maximum value, w, of scene completion speed index in all solutions output by the algorithm1、w2、w3、w4Respectively, the importance weights of the corresponding targets according with the preference of the decision maker.
4. The method of claim 2, wherein the improved enhanced pareto evolutionary algorithm ISPEA2 multi-objective optimization algorithm is:
step 3.1: the method comprises the steps of carrying out population coding, initializing a population, wherein the population coding adopts one-dimensional sequential coding; one locus in each chromosome represents one of the parameters to be optimized;
step 3.2: measuring the population diversity D (P) if the population diversity is less than a fixed threshold dlowThen step 3.7 is executed; otherwise, executing step 3.3;
step 3.3: selecting chromosomes in the population by adopting a non-dominance rule-based method according to a mechanism of strengthening the pareto SPEA2 algorithm;
before each iteration is started, an external population pool is set for generating a population entering the next generation, and the external population pool of the t generation is set as
Figure FDA0003474671980000041
The t generation is composed of external population pool
Figure FDA0003474671980000042
The resulting population is ptTo generate an external population pool of the t +1 generation
Figure FDA0003474671980000043
Firstly, the following components are mixed
Figure FDA0003474671980000044
And ptAll non-dominant solution additions in
Figure FDA0003474671980000045
Then, to
Figure FDA0003474671980000046
Judging the size of the product; if it is not
Figure FDA0003474671980000047
The number of chromosomes in the chromosome is greater than a predetermined number
Figure FDA0003474671980000048
Then reduced by population density based methods
Figure FDA0003474671980000049
Up to the number of chromosomes in
Figure FDA00034746719800000410
Number of chromosomes in is equal to
Figure FDA00034746719800000411
If the number of chromosomes in the population is less than
Figure FDA00034746719800000412
Then through a population density based selection method will
Figure FDA00034746719800000413
And ptThe solution in the dominant solution is added according to the order of the dominant number from small to large
Figure FDA00034746719800000414
Up to
Figure FDA00034746719800000415
Number of chromosomes in is equal to
Figure FDA00034746719800000416
Step 3.4: performing cross operation on the population; randomly swapping a part of the gene locus between two chromosomes;
step 3.5: carrying out mutation operation on the population; randomly selecting at most three gene positions in a chromosome, and randomly generating other numerical values to replace numerical values on the gene positions in the selected chromosome so as to realize the variation of the chromosome;
step 3.6: judging whether a termination condition is reached, if not, collecting the known pareto solution of the current generation and returning to the step 3.2, otherwise, finishing the algorithm and outputting a result;
step 3.7: generating artificial chromosomes from chromosomes stored in an external storage pool;
step 3.8: and replacing the chromosome with smaller contribution degree to population entropy in the current population by the artificial chromosome and executing the step 3.4.
5. A method according to claim 3, wherein the measure of population diversity in step 3.2 is:
Figure FDA00034746719800000417
ps is the number of chromosomes in the population, sijIs the value at the jth gene position on the ith chromosome in the population,
Figure FDA00034746719800000418
is the mean value of the jth locus on all chromosomes in the whole population, | L | is a solution space pairThe length of the corner line.
6. The method of claim 4, wherein the method of generating artificial chromosomes from chromosomes stored in the external storage pool in step 3.7 is:
step 3.7.1: calculating the contribution degree of each chromosome in the current population to the current population entropy,
first, the probability of the value v appearing on the chromosome locus j in the current population is calculated:
Figure FDA0003474671980000051
wherein, PvjIs the probability that the value v appears at locus j, XvjThe number of times that the value v is distributed to the gene position j in all the chromosomes of the current population after statistics;
secondly, calculating the contribution degree of the gene position j on the chromosome to the population entropy;
Figure FDA0003474671980000052
v is the complete set of values V;
step 3.7.2: calculating the entropy of the whole population;
Figure FDA0003474671980000053
wherein the PCDpAs entropy of the entire population, NpIs the dimension of the problem;
step 3.7.3: calculating the contribution degree of each chromosome i in the current population to the entropy of the whole population,
CD(i)＝PCDp-PCDp-i
CD (i) degree of contribution of chromosome i to entropy of whole population, PCDpFor entropy of the entire population, PCDp-iEntropy of the population after chromosome i has been removed for the entire population;
step 3.7.4: chromosomes used to generate artificial chromosomes are collected in an external storage pool: collecting the non-dominant solution set generated in the step at each step of iteration and adding the solution set into an external storage pool, when the number of chromosomes in the external storage pool is less than a preset threshold value
Figure FDA0003474671980000054
When all non-dominant solutions are placed in the external pool, when the number of chromosomes in the external pool is greater than that
Figure FDA0003474671980000055
And sequencing the chromosomes in the external storage pool according to the contribution degree of the chromosomes to the population entropy, and removing the individuals with low contribution degree to the population entropy out of the storage pool until the number of the chromosomes in the external storage pool is equal to that in the external storage pool
Figure FDA0003474671980000056
Step 3.7.5: and respectively counting the probability of each numerical value v appearing on each gene position of all chromosomes in the external storage pool, and assigning values to each gene position on the artificial chromosome according to the probability value when the artificial chromosome is generated until all required artificial chromosomes are generated.
7. A multi-target control parameter tuning system of an unmanned aerial vehicle cluster control system, comprising a memory and a processor, wherein the memory stores a computer program of the multi-target control parameter tuning method of the unmanned aerial vehicle cluster control system, and the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.
CN202110005312.1A 2021-01-04 2021-01-04 Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system Active CN112783210B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110005312.1A CN112783210B (en) 2021-01-04 2021-01-04 Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110005312.1A CN112783210B (en) 2021-01-04 2021-01-04 Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system

Publications (2)

Publication Number Publication Date
CN112783210A CN112783210A (en) 2021-05-11
CN112783210B true CN112783210B (en) 2022-03-25

Family

ID=75755246

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110005312.1A Active CN112783210B (en) 2021-01-04 2021-01-04 Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system

Country Status (1)

Country Link
CN (1) CN112783210B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114252785B (en) * 2021-12-13 2024-02-27 深圳市道通智能航空技术股份有限公司 Unmanned aerial vehicle control method, electronic equipment and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007001962A2 (en) * 2005-06-20 2007-01-04 Kurzweil Technologies, Inc. Systems and methods for generating biological material
EP1901153A1 (en) * 2006-09-12 2008-03-19 OFFIS e.V. Control system for unmanned 4-rotor-helicopter
CN107169608A (en) * 2017-05-27 2017-09-15 合肥工业大学 Multiple no-manned plane performs the distribution method and device of multitask
CN108196583A (en) * 2017-08-21 2018-06-22 中国人民解放军陆军工程大学 Unmanned plane cluster control method
CN108459612A (en) * 2017-02-21 2018-08-28 北京航空航天大学 Unmanned plane formation control method based on Artificial Potential Field Method and device
CN109445459A (en) * 2018-10-25 2019-03-08 中国人民解放军战略支援部队航天工程大学 Unmanned plane cluster formation control method and its device
CN111598473A (en) * 2020-05-22 2020-08-28 中国人民解放军国防科技大学 Multi-platform joint task planning method for complex observation task
CN112000131A (en) * 2020-09-09 2020-11-27 中国人民解放军国防科技大学 Unmanned aerial vehicle cluster path planning method and system based on artificial potential field method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10244675B2 (en) * 2016-01-29 2019-04-02 Veritas FC LLC Field robot

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007001962A2 (en) * 2005-06-20 2007-01-04 Kurzweil Technologies, Inc. Systems and methods for generating biological material
EP1901153A1 (en) * 2006-09-12 2008-03-19 OFFIS e.V. Control system for unmanned 4-rotor-helicopter
CN108459612A (en) * 2017-02-21 2018-08-28 北京航空航天大学 Unmanned plane formation control method based on Artificial Potential Field Method and device
CN107169608A (en) * 2017-05-27 2017-09-15 合肥工业大学 Multiple no-manned plane performs the distribution method and device of multitask
CN108196583A (en) * 2017-08-21 2018-06-22 中国人民解放军陆军工程大学 Unmanned plane cluster control method
CN109445459A (en) * 2018-10-25 2019-03-08 中国人民解放军战略支援部队航天工程大学 Unmanned plane cluster formation control method and its device
CN111598473A (en) * 2020-05-22 2020-08-28 中国人民解放军国防科技大学 Multi-platform joint task planning method for complex observation task
CN112000131A (en) * 2020-09-09 2020-11-27 中国人民解放军国防科技大学 Unmanned aerial vehicle cluster path planning method and system based on artificial potential field method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
A new consensus theory-based method for formation control and obstacle avoidance of UAVs;Yu Wu;《Aerospace Science and Technology》;20201105;全文 *
An intelligent cooperative mission planning scheme of UAV swarm in uncertain dynamic environment;Ziyang Zhen;《Aerospace Science and Technology》;20200306;全文 *
智能机器人可变参数群体控制模型的多目标优化方法;王原;《中国科学:技术科学》;20200531;全文 *
面向无人机集群路径规划的智能优化算法综述;杨 旭;《控制理论与应用》;20201130;全文 *

Also Published As

Publication number Publication date
CN112783210A (en) 2021-05-11

Similar Documents

Publication Publication Date Title
Han et al. Adaptive gradient multiobjective particle swarm optimization
Zeebaree et al. Combination of K-means clustering with Genetic Algorithm: A review
Kaur et al. A novel multi-objective bacteria foraging optimization algorithm (MOBFOA) for multi-objective scheduling
Hu et al. Many-objective particle swarm optimization using two-stage strategy and parallel cell coordinate system
Saif et al. Multi-objective artificial bee colony algorithm for simultaneous sequencing and balancing of mixed model assembly line
Jiang et al. Research on large-scale bi-level particle swarm optimization algorithm
CN107886201B (en) Multi-objective optimization method and device for multi-unmanned aerial vehicle task allocation
CN105929690B (en) A kind of Flexible Workshop Robust Scheduling method based on decomposition multi-objective Evolutionary Algorithm
Rashno et al. Particle ranking: An efficient method for multi-objective particle swarm optimization feature selection
Colby et al. Shaping fitness functions for coevolving cooperative multiagent systems.
Coello Coello Evolutionary multi-objective optimization: basic concepts and some applications in pattern recognition
Qian et al. A multi-objective evolutionary algorithm based on adaptive clustering for energy-aware batch scheduling problem
Long et al. Clustering stock data for multi-objective portfolio optimization
CN112783210B (en) Multi-target control parameter optimization method of unmanned aerial vehicle cluster control system
Koduru et al. Fuzzy dominance based multi-objective GA-simplex hybrid algorithms applied to gene network models
Lu et al. A Population Cooperation based Particle Swarm Optimization algorithm for large-scale multi-objective optimization
Tan et al. Multi-UAV path planning based on IB-ABC with restricted planned arrival sequence
Xin et al. An estimation of distribution algorithm for multi-robot multi-point dynamic aggregation problem
Cheung et al. Hierarchical particle swarm optimizer for minimizing the non-convex potential energy of molecular structure
Zhang et al. A parallel hybrid evolutionary particle filter for nonlinear state estimation
Liu An analysis of particle swarm optimization of multi-objective knapsack problem
CN113141272B (en) Network security situation analysis method based on iteration optimization RBF neural network
Hashemi Borzabadi et al. Approximate Pareto optimal solutions of multi objective optimal control problems by evolutionary algorithms
CN111860755A (en) Improved particle swarm algorithm based on regression of support vector machine
Cao et al. Cluster analysis based on attractor particle swarm optimization with boundary zoomed for working conditions classification of power plant pulverizing system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant